A real challenge with the availability of data this year has been a) public availability of data has lead to substantially all the distributed science we're seeing, saving lives and b) the "just give me one number" stats are misinterpreted by many, *including by policy makers.*
The count, the count, the count. Today's count? Lower than yesterday's? Amazing, we're winning.
The count is *an imperfect proxy* about *one particular dimension of the problem* which *in universes where things are bad will lie to us.*
The count is *an imperfect proxy* about *one particular dimension of the problem* which *in universes where things are bad will lie to us.*
Even if you had a magical wand which accurately reported the count every day with no reporting delay (and that would not be the first magical wand you would buy!), the count being down doesn't mean to be optimistic!
Flow/stock problem compared to flow/stock problem on beds/etc.
Flow/stock problem compared to flow/stock problem on beds/etc.
And the count, the count, the count is, after the epidemic starts, generally lower than the marginal amount of capacity you can bring online in a day.
If you go over the capacity number, then patients overflow into general medical system.
And then, the count starts going up.
If you go over the capacity number, then patients overflow into general medical system.
And then, the count starts going up.
So you would hope that the daily announcement would not be the count. The daily announcement should be "We continue fighting the crisis. We urgently call on the citizenry to observe latest guidance. We are at X% capacity, which is *worse than yesterday*. Please redouble efforts."
* generally *not* lower
